“Big Data” is the buzzword of the moment. However, what is Big Data and does it have any relevance for clinicians working on the front lines, or is it simply a term that applies to executives in health systems or large organizations? Wikipedia defines Big Data as “a collection of data sets so large and complex that it becomes difficult to process using on-hand database management tools. The challenges include capture, curation, storage, search, sharing, analysis, and visualization.” There are many different definitions, each with a slightly different nuance. At the core it is the creation of massive amounts of data through tools such as social media, monitors, or sensors that are continuously generating information. Healthcare is data intensive. Billions of data elements are generated every day through diagnostics, vital signs measurements, documentation, genetic analysis, and so on. The challenge is making this information useful to individuals who provide care as well as those who manage and monitor the system.
One of the ways to make Big Data more useful is to organize it in a way that it is easily searchable or comparable. In other words, based upon some standards. Elements that are the same as one another can more easily be compared, monitored, and shared. However, historically this has not been the way that data is generated or collected. For example, many EHR vendors initially developed proprietary methods to collect and catalog information in their systems. Generally this happened for both tactical and strategic reasons. In some instances, there were no nationally recognized standards that were available and, as a result, vendors developed their own set of standards to store and retrieve data. The more strategic reason was that proprietary standards provided a company with a mechanism to hold more tightly onto their customers. It was not easy to change from one system to another if the information could not easily be transferred. In 2012, there are a number of healthcare standards organizations. One that stands out as a leader is HL7 International. Last week I had the opportunity to talk with Bob Dolin, president and chief medical officer of Lantana Consulting Group and chair elect of HL7 International. The question that I posed to Bob was how to make standards more clinically relevant. Listen to the podcast.
Bob talked about a different approach to the management of Big Data in healthcare. He offered a very thoughtful analogy. When Google began to catalog the World Wide Web, it did not require every site to have the same 20 elements in Google standard format; it basically took every piece of data it could get and slowly began the process of cataloging that information using spiders and algorithms to identify and make the information meaningful. Over time, Google improved and developed news way to collect and store data that made the disorganized Web extremely user friendly. It still amazes me that one can search any term and be provided with relevant information in a fraction of a second from vast amounts of data.
Bob’s suggestion is to approach Big Data by applying incremental structure. He recommends beginning with clinically relevant standardized clinical (CDA) documents that providers need in order to deliver care and sharing these documents. Over time, further structure is added and the selection of clinical documents is expanded providing greater value to clinicians. This process mirrors Google’s approach in terms of making the data more interoperable between systems and providers.
There is too much information to simply stop, standardize, and organize the data and then press start all over again. We need mechanisms to manage this information while the healthcare leviathan is in motion without stopping or slowing down. Big Data with incremental structure could be just the answer.
Do you agree or disagree? Add your thoughts below.